Hello folks,
I have an interview call later this week which the work is regarding implementing generative AI within the companies workflow. Using LLMs with finetuning/in-context learning using system logs etc kind of stuff.
I have studied machine learning, worked for few years now as well. Have good understanding of those stuff but never tried fine tuning hands-on. I’m worked majority into computer-vision applications but think that I lagged a bit on the LLM side.
Any suggestions, recommeded papers, courses, videos I could go through?
Thanks!
You must log in or register to comment.
Not an exhaustive list, but here’s a few suggestions:
- Get really familiar with embedding/semantic-search/RAG.
- Fine-tune a LLaMA-2 7B model using QLoRA on an A10 EC2 instance (or whatever compute you have) to do something like document classification or sentiment analysis.
- Watch this video: https://www.youtube.com/watch?v=yj-wSRJwrrc
- Read this paper: https://arxiv.org/pdf/2311.04235.pdf
Thanks alot. Surely helpful
Good luck with your interview
https://rentry.org/llm-training
Challenges and Applications of Large Language Models
https://github.com/bkitano/llama-from-scratch
https://www.philschmid.de/tags/generativeai
https://web.stanford.edu/class/cs224n/ and 224u, 224v
https://eugeneyan.com/writing/llm-patterns/ (many other good blogposts)
https://huyenchip.com/2023/04/11/llm-engineering.html
RAG stuff(sentence embedding, vector db):
https://www.pinecone.io/learn/
https://www.sbert.net/
https://haystackconf.com/us2023/keynote/
https://www.latent.space/p/llamaindex#detailsAn actual saint
I put together an overview of fine-tuning recently, focusing on some more practical techniques when you really get into the process. https://david010.medium.com/fine-tuning-llms-practical-techniques-and-helpful-tips-3a169cc62cca
Thanks for this. Super useful
what level of education are you? PhD?
Take a look at TensorGym, which offers practical hands-on ML coding exercises (disclosure: I am building TensorGym)
Good luck with the interview!Thanks! Bombed that interview but will try it for my next one!